AI medical coding promises efficiency gains, but most implementations cannot deliver the desired ROI in production. Standard methods optimize for benchmarks that do not reflect the complexity of real billing workflows, undermining business KPIs. This session examines why that gap exists and what it takes to close it. Through realistic simulation-based examples, you will learn how pairing automation with risk-aware review enables human coders to focus their expertise where it delivers the greatest financial impact.

[ A note from event chair Eric Siegel: This approach is “hybrid AI” in that one system may automatically code medical insurance claims, and another then predicts the risk of a claim’s subsequent denial, thus enacting the predictive reliability layer that I refer to in my opening keynote, “Predictive AI’s New Killer App,” and that I cover in this article. ]